---
title: Models that can read your images
description: The 18 of 60 models Else can run where we have measured image input working — hand them a screenshot or a photo and they read it.
---

{/*
  ============================================================================
  DO NOT EDIT THIS FILE BY HAND. IT IS GENERATED, AND YOUR EDIT WILL BE LOST.

  Generated by : apps/platform/scripts/build-docs-model-guides.ts
                 (in the chatinone-platform repo)
  Generated from: packages/core/data/models.json  ->  pickerModels(CATALOG_MODELS)
                 apps/platform/src/model-routing.ts  (DEFAULT_TIER_BINDINGS)
  This page     : read-your-images

  To change anything below, change the source in that repo and re-run:
      pnpm --filter @chatinone/platform gen:docs-model-guides -- --write

  Committed snapshots live in that repo under
  apps/platform/test/fixtures/docs-model-guides/, and
  apps/platform/test/docs-price-table-drift.test.ts fails if a snapshot, or the
  SET of pages, disagrees with the regenerated output.

  Prices here are RETAIL — what you are actually charged. They come from the
  same projection the in-product engine picker is built from, so this page and
  the product cannot disagree.
  ============================================================================
*/}

If the material you want to hand over is a picture — a screenshot of a dashboard, a photo of a receipt, a scan of a form — you need a model that can actually look at it. 18 of the 60 models Else can run have had image input **measured working**. These are those 18.

## The models

| Model | Runs as | Context | Input / 1M tokens | Output / 1M tokens |
| --- | --- | --- | --- | --- |
| `gemini-2.5-flash-lite` | Exact engine pick | 1M | $0.12 | $0.48 |
| `gpt-4o-mini` | Exact engine pick | 128K | $0.18 | $0.72 |
| `gemini-3.1-flash-lite` | Exact engine pick | 1M | $0.30 | $1.80 |
| `llama-4-scout-17b-16e-instruct` | Exact engine pick | 131K | $0.32 | $1.02 |
| `mistral-small-3.1-24b-instruct` | Exact engine pick | 128K | $0.42 | $0.67 |
| `qwen3.5-397b-a17b` | Exact engine pick | — | $0.72 | $4.32 |
| `gpt-5.4-mini` | Exact engine pick | 128K | $0.90 | $5.40 |
| `claude-haiku-4.5` | Exact engine pick | 200K | $1.20 | $6.00 |
| `gemini-2.5-pro` | Exact engine pick | 1M | $1.50 | $12.00 |
| `gpt-5.1` | Exact engine pick | 128K | $1.50 | $12.00 |
| `grok-4.3` | Exact engine pick | 1M | $1.50 | $3.00 |
| `gemini-3.5-flash` | Exact engine pick | 1M | $1.80 | $10.80 |
| `gpt-4.1` | Exact engine pick | 1M | $2.40 | $9.60 |
| `gpt-4o` | Exact engine pick | 128K | $3.00 | $12.00 |
| `gpt-5.4` | Exact engine pick | 1M | $3.00 | $18.00 |
| `claude-sonnet-4.5` | Exact engine pick | 200K | $3.60 | $18.00 |
| `claude-opus-5` | **Max** tier default | 1M | $6.00 | $30.00 |
| `gpt-5.5` | Exact engine pick | 1M | $6.00 | $36.00 |

Every price above is paired with how you reach that model: a named tier runs its model by default, and everything else is reached from **Pick an exact engine** in the composer's engine menu.

## How this list is decided

We probe each model's image input ourselves and record the result as one of three states: measured working, measured not working, or not measured. This page lists only the first state. The probe result is the same fact the **Images in** column shows on [Model prices](/docs/model-prices).

Measured working means a real probe got a real answer about a real image. It does **not** mean the model reads every image equally well, and it is not a claim about accuracy on your particular picture — we publish no measured quality ranking. If a model reads your screenshot badly, try another on this list rather than assuming the list is wrong.

## Next

- [Model prices](/docs/model-prices) — all 60 models, all prices, one table
- [The Fast tier](/docs/tiers/fast), [The Smart tier](/docs/tiers/smart), [The Max tier](/docs/tiers/max) — what you get without choosing a model at all
- Start a run at [elsedoes.com/home](https://elsedoes.com/home)

**Prices last changed 2026-07-29** (source `ced6a4f`).